Insight Into ICH and Warfarin Dynamics May Lead to Safer Anticoagulation Protocols

To kick-start his presentation on reversal of coagulopathy during a symposium at the International Stroke Conference (ISC) that took place in San Francisco in early February, J. Claude Hemphill III, MD, professor of neurology at the University of California, San Francisco, displayed a CT scan illustrating a modest intracerebral bleed in the brain of a 64-year-old woman who had been taking warfarin for the treatment of atrial fibrillation. Hemphill noted that standard care, including administration of vitamin K and fresh frozen plasma (FFP), was given.

To kick-start his presentation on reversal of coagulopathy during a symposium at the International Stroke Conference (ISC) that took place in San Francisco in early February, J. Claude Hemphill III, MD, professor of neurology at the University of California, San Francisco, displayed a CT scan illustrating a modest intracerebral bleed in the brain of a 64-year-old woman who had been taking warfarin for the treatment of atrial fibrillation. Hemphill noted that standard care, including administration of vitamin K and fresh frozen plasma (FFP), was given.

"She was doing okay," Hemphill remarked. He then showed what the patient's brain looked like 10 hours later. The audience-a-the majority of whom were presumably stroke specialists-gasped. The left hemisphere was pretty much wiped out by the bleed.

Persons receiving warfarin reportedly have a small annual risk-about 0.2% to 0.6%-of intracerebral hemorrhage (ICH),1 but this reportedly can climb up to 1.8% in stroke-prone patients.2 And when warfarin-associated ICH occurs, it is often massive-and fatal. "It's not only that you bleed, you have severe hemorrhage," remarked Jonathan Rosand, MD, assistant professor of neurology at Harvard Medical School during his presentation at the ISC symposium. Rosand, Hemphill, and others spoke about ways to prevent and troubleshoot the problem.

Warfarin is highly effective in managing atrial fibrillation and thus in ameliorating stroke incidence. "Dose-adjusted warfarin is associated with a 68% relative risk reduction in stroke compared with no treatment and a 52% reduced risk compared with aspirin," noted Rosand, citing unpublished research (scheduled to appear in the American Journal of Medicine later this year) by a team led by Margaret Fang, MD, MPH, assistant adjunct professor of medicine at the University of California, San Francisco Medical Center.

The gravity of warfarin-associated ICH, however, is such that it can make warfarin use seem dicey. To elucidate what is known about warfarin and the risk of hemorrhage in general, Rosand cited research by Fang and colleagues3 that analyzed data on 13,559 persons with nonvalvular atrial fibrillation.

"The team looked at the risk of major hemorrhages on warfarin, analyzing 15,300 person-years. They found 170 major hemorrhages. Of these, 72 were intracranial and 98 were major extracranial. Strikingly, 76% of the intracranial hemorrhagic events resulted in severe disability or death," Rosand reported.

He then showed data from his own institution.4 "We looked at outcomes at 90 days in 400 patients with ICH who passed through our emergency department." Patients were divided into 2 groups depending on whether they were receiving warfarin at the time of presentation. Death, severe disability, moderate disability, and full recovery were recorded (Table 1). Notably, the rates of death and severe disability for patients receiving warfarin were double those of patients who were not. Whereas factors such as hyperglycemia affected initial volume of the hemorrhagic lesion, warfarin was the sole factor implicated in expansion of the lesion, resulting in higher mortality among affected patients. But there is more to the story.

 Table 1 - Outcomes (%) at 90 days of patients with ICH
Taking warfarin Death Severe disability Moderate disability Full recovery 
No 26.3 42.5 16.2 15 
Yes 52 31.4 8.8 7.8
ICH, intracerebral hemorrhage.

In the previously cited study by Fang and colleagues,3 it was revealed that although 170 major hemorrhages occurred during 15,300 person-years of warfarin therapy, almost as many-162 major hemorrhages-occurred in the slightly larger time-window of 15,530 person-years of non-warfarin exposure. Furthermore, hemorrhage rates rose with age, with those patients aged 80 years and older being particularly vulnerable. This trend was independent of warfarin use, leading the study authors to conclude that older age is a major risk factor for ICH in patients with atrial fibrillation regardless of exposure to warfarin. They concluded that warfarin could be safely used in carefully selected older patients.

Rosand also reported that length of time on warfarin therapy has an impact on risk. "The overwhelming majority of patients taking warfarin are having their ICH after a year of warfarin therapy," he declared, pointing to interim data from the Genetic Risks for Medication-Related Hemorrhagic Stroke (also known as GOCHA) trial being conducted through the NIH and the National Institute of Neurological Diseases and Stroke.

Rosand, a GOCHA trial investigator, explained that when the length of time patients were receiving warfarin was analyzed, a trend emerged (Table 2). Seventy-one percent of patients who were taking warfarin when an ICH occurred had been taking it for at least a year.

 Table 2 - Incidence (%) of ICH over time
  < 3 mo 3 mo to 1 y  > 1 y  
Patients taking warfarin (n = 187) 8 21  71  
Controls (n = 167) 3 42  55 
ICH, intracerebral hemorrhage.

"Warfarin-related ICH doesn't appear to be strongly related to the early period of warfarin use, when dosing variability might be an issue," said Rosand.


In a study published in Neurology in 2000, Rosand and colleagues1 demonstrated a link between warfarin- associated ICH, cerebral amyloid angiopathy (CAA), and the apolipoprotein E (APOE) ε2 allelle. Their aim at the outset of the study was to show that assuming an association between CAA and warfarin-associated ICH was confirmed, affected patients would be more likely to have increased frequencies of APOE €2 or ε4 alleles than controls who were receiving warfarin but in whom ICH had not occurred.

Indeed, while APOE ε3 and ε4 frequencies were comparable between controls and those in whom hemorrhages had occurred, the APOE ε2 frequency was 0.11 in patients in whom an ICH occurred versus 0.04 in controls. Further, the higher frequency was specifically seen in a subgroup of affected patients who had lobar hemorrhages as opposed to deep hemorrhages, owing, in part, to the fact that the deeper brain is not as vascularly rich as the lobar regions. Rosand and colleagues concluded that warfarin-associated hemorrhages may be attributable to predisposing vascular factors more than excessive anticoagulation.

CAA contributes to leakiness of blood vessel walls, resulting in microhemorrhages. "Warfarin then causes these little hemorrhages to become big hemorrhages," Rosand remarked. Although patients are not (yet) able to be screened for CAA before a hemorrhage occurs, insight into the link between CAA and ICH may guide patient selection and warfarin use in the future, according to Rosand and his team.

In a follow-up study,5 Rosand and fellow investigators confirmed a link between white matter disease and warfarin-associated ICH in survivors of ischemic stroke, which was first reported in the Stroke Prevention in Reversible Ischemia Trial (SPIRIT).6 Twenty-four (92%) of 26 patients in whom an ICH had occurred had white matter disease, compared with 27 (48%) of 56 controls. Rosand noted that other studies7,8 have suggested that like CAA, white matter disease may be heritable.

"Effective therapies for small-vessel disease may reduce the risk of anticoagulant hemorrhage. Identifying genes may help identify those at risk for anticoagulant-type hemorrhage," Rosand concluded.


Hemphill noted that his 64-year-old patient did not survive the massive hemorrhage that developed within 10 hours of hospital admission. He went on to note that standard recommendations for emergency intervention are not so great.

"When we are reversing warfarin, what's the usual way? We transfuse 2 units of FFP and give some vitamin K. That's okay-but not so fast. Does it really work?" questioned Hemphill. He cited a recent study by Goldstein and colleagues9 showing that the earlier FFP is given, the more likely the international normalized ratio (INR) would be corrected. However, the study also found that rapid correction of the INR did not reduce morbidity or mortality. The study authors postulated that an even earlier and more rapid means of anticoagulation reversal is required.

"This sets up a challenge for us," said Hemphill. "There are 3 different options in anticoagulation reversal: directly compete with the vitamin K antagonists, replace the native coagulation factors with FFP, or use prothrombin complex concentrate [PCC] to bypass the coagulation cascade and give recombinant activator factor VII [rFVIIa]." He noted that most centers in the United States rely on FFP and vitamin K, although other possibly more efficient, albeit more expensive, options are available.

"The plasma half-life of vitamin K is 1½ to 3 hours, but the full effect occurs 24 hours after administration. The recommended dose is 5 to 10 mg," explained Hemphill. "A 10-mg dose is cheap," he noted, at $4.77 for a 10-mg ampule. Although it doesn't have good utility for urgent warfarin reversal in patients with acute ICH, it should be given adjunctively to offset the short half-lives of other agents used for anticoagulant reversal, according to Hemphill-a recommendation echoed by others.2

Although FFP administration is the standard approach, it requires compatibility testing, which imposes a delay in administration. Further, a substantial amount may be required to correct the INR, said Hemphill. "Our usual size of FFP is 200 mL. Our blood bank charges us $72 for that. It is still pretty cheap," he commented, adding, "Its utility for warfarin reversal is poor in relation to logistics in relation to time. Should it still be given? Probably."

PCC is gaining recognition as a more efficient acute therapy, although it is very costly. "It takes 20 minutes to reconstitute. No compatibility testing is required. Its utility in warfarin reversal is pretty good, but there are problems with availability in some parts of the United States, and the agent carries the potential risk of causing thrombosis," said Hemphill.

The agent contains coagulation factors VII, IX, and X; prothrombin; and proteins C, S, and Z.2 "Available formulations differ slightly in relation to the amount of factor VII or factor IX and also vary in the amount to be given: often 30 to 50 IU in a single bolus, which ends up being 50 mL in volume. For a 70-kg person who gets a 50-IU dose in a hospital in San Francisco, the cost is $2380-so it's quite expensive," Hemphill commented.

rFVIIa, an agent used to control bleeding in hemophilia, has shown some promise in other forms of uncontrollable bleeding and may be a useful adjunctive therapy in anticoagulation reversal interventions, but more research is needed, according to Hemphill, and it is very expensive, he added. Nevertheless, guidelines10 emerging from the Seventh American College of Chest Physicians' Conference on Antithrombotic and Thrombolytic Therapy recommended PCC and rFVIIa supplemented by vitamin K for life-threatening bleeding and elevated INR, noted Hemphill.


Reinstituting anticoagulation therapy in patients with atrial fibrillation who have survived lobar ICH is not a good idea-at least not on a long-term basis, reported Mark H. Eckman, MD, professor of medicine at the University of Cincinnati. He went through the literature and then reported on findings from his own research,11 published in Stroke in 2003, that showed that patients in whom anticoagulation therapy was not reinstituted could expect more "quality-adjusted life-years" (5.4 vs 3.5) than patients in whom it was reinstituted at some point after the index lobar ICH.

An alternative in selected patients might be aspirin therapy, according to the research by Eckman and colleagues.11 Their analysis, however, showed that aspirin was a relatively safe alternative for patients who were at higher risk for ischemic stroke than ICH. Nevertheless, a recent prospective cohort study by Viswanathan and colleagues12 suggested that antiplatelet therapy is common after ICH and not associated with a substantial increase in ICH recurrence.

Eckman also stressed that screening methods that include imaging for identification of microbleeds and genetic testing for APOE ε2 carriage and other known risk factors should be considered so that careful patient selection can be applied to anticoagulant therapy.




Rosand J, Hylek EM, O'Donnell HC, Greenberg SM. Warfarin-associated hemorrhage and cerebral amyloid angiopathy: a genetic pathologic study.




Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: current practices and unresolved questions.




Fang MC, Go AS, Hylek EM, et al. Age and the risk of warfarin-associated hemorrhage: the anticoagulation and risk factors in atrial fibrillation study.

J Am Geriatr Soc.



Flibotte JJ, Hagan N, O'Donnell J, et al. Warfarin, hematoma expansion, and outcome of intracerebral hemorrhage.




Smith EE, Rosand J, Knudsen KA, et al. Leukoaraiosis is associated with warfarin-related hemorrhage following ischemic stroke.


2002;59: 193-197.


Gorter JW. Major bleeding during anticoagulation after cerebral ischemia: patterns and risk factors. Stroke Prevention In Reversible Ischemia Trial (SPIRIT). European Atrial Fibrillation Trial (EAFT) study groups.




DeStefano AL, Atwood LD, Massaro JM, et al. Genome-wide scan for white matter hyperintensity: the Framingham Heart Study.




Turner ST, Jack CR, Fornage M, et al. Heritability of leukoaraiosis in hypertensive sibships.




Goldstein JN, Thomas SH, Frontiero V, et al. Timing of fresh frozen plasma administration and rapid correction of coagulopathy in warfarin-related intracerebral hemorrhage.




Ansell J, Hirsh J, Poller L, et al. The pharmacology and management of the vitamin K antagonists: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy [published correction appears in


2005;127: 415-416].


2004;126(3 suppl):204S-233S.


Eckman MH, Rosand J, Knudsen KA, et al. Can patients be anticoagulated after intracerebral hemorrhage? A decision analysis.




Viswanathan A, Rakich SM, Engel C, et al. Antiplatelet use after intracerebral hemorrhage.